Ecificity was 79.4 , false positive rate () was 20.6 , false negative rate () was 25.9 , positive predictive value (PPV) was 76.8 , negative predictive value (NPV) was 76.9 , positive likelihood ratio (+LR) was 3.59,Figure 2 Correlation coefficients between two variables of pre-selected variables.Gao et al. BMC Medical Informatics and Decision Making 2012, 12:58 http://www.biomedcentral.com/1472-6947/12/Page 6 ofTable 5 Univariate logistic regression analysis of the 12 pre-selected serological parametersParameter FIB D2 sIgA sIgG UN ALB TG CH DB ALP CA199 CA153 B 2.485 -0.242 0.009 0.001 0.049 0.109 -0.326 -0.429 0.308 -0.007 -0.041 -0.071 S.E. 0.735 0.161 0.002 0.001 0.057 0.026 0.186 0.131 0.126 0.005 0.019 0.031 Wald 11.432 2.275 16.440 6.396 0.740 17.498 3.068 10.673 5.970 1.883 4.855 5.293 df 1 1 1 1 1 1 1 1 1 1 1 1 Sig. 0.001 0.132 0.000 0.011 0.390 0.000 0.080 0.001 0.015 0.170 0.028 0.021 Exp(B) 0.534 0.785 1.009 1.001 1.051 1.115 0.722 0.651 1.361 0.993 0.960 0.Abbreviations: Omeprazole sodium FIB, fibrinogen; D2, D-dimer; sIgA, serum immunoglobulin A; sIgG, serum immunoglobulin G; UN, urea; ALB, albumin; TG, triglyceride; CH, total cholesterol; DB, direct bilirubin; ALP, alkaline phosphatase; CA199, carbohydrate antigen; CA153, carbohydrate antigen 15?.negative PubMed ID:https://www.ncbi.nlm.nih.gov/pubmed/16570919 likelihood ratio (-LR) was 0.32, and Youden's index was 0.535. The area under the ROC curve with PRE-1 for IgAN was 83.8 (P < 0.0001, 95 CI: 0.7660.910) (Figure 3). The "FIB + sIgA + Manifestation" combination was significant in the classification of IgAN and non-IgAN, as determined via logistic regression analysis. The classification equation, which includes these 3 parameters, for predicting IgAN is as follows:PRE ?1 ?1 ?1 ? ?e 0:648?:326FIB?:011sIgA?:089Manifestation?Figure 3 Figure 3 Area under ROC curve of the predicted probability of IgAN with "FIB + sIgA + Manifestation" combination from logistic regression. Area under the ROC curve for predicting immunoglobulin A nephropathy (IgAN) with the equation derived via logistic regression analysis, which includes the "fibrinogen (FIB) + serum immunoglobulin A level (sIgA) + manifestation" combination. The state variable is IgAN.was 76.9 , sensitivity was 79.3 , specificity was 74.6 , false positive rate () was 25.4 , false negative rate () was 20.7 , positive predictive value (PPV) was 74.2 , negative predictive value (NPV) was 79.7 , positive likelihood ratio (+LR) was 3.12, negative likelihood ratio (-LR) was 0.28, and Youden's index was 0.461. Theb) Model based on discriminant analysis: The 13 preselected parameters were substituted into a step discriminant analysis. The predicted probabilities (PRE-2) were calculated and saved. Similar to the logistic regression analysis, only sIgA, manifestation, and FIB were significant in the classification of IgAN and non-IgAN (Table 7). The classification model with these 3 parameters was evaluated, and it was found that accuracyTable 6 Parameters used in the multivariate logistic regression analysis for developing the classification modelParameter B S.E. Wald df Sig. Exp 95 CI for EXP(B) (B) Lower Upper 0.147 0.522 1.006 0.771 0.997 1.Table 7 Parameters used in the discriminant analysis for developing the classification modelVariables in the analysis Step 1 2 ALB ALB sIgA 3 ALB sIgA Manifestation 4 sIgA Manifestation 5 sIgA Manifestation FIB Tolerance 1.000 1.000 1.000 0.601 0.974 0.593 0.986 0.986 0.967 0.803 0.789 Sig. of F to Remove 0.000 0.000 0.000 0.123 0.000 0.020 0.000 0.000 0.000 0.0.